DE10060285A1 - Method for operating a control arrangement for a hydraulic system - Google Patents

Abstract

The invention relates to a method for operating a control device for a hydraulic system, according to which the actual values in said hydraulic system are detected by sensors (2) and in particular the pressure (p) of the hydraulic fluid is controlled. Said method is characterised in that the effective system volume of hydraulic fluid in the sealed volume of the pressure-controlled system is calculated using the measured values and that system parameters for the parameterisation of the control device are derived from the calculated system volume.

Description

State of the art

The invention relates to methods for operating a Control arrangement and such a control arrangement for a hydraulic system according to the genus of the main claim.

It is already one from DE 198 28 752 A1 Known control arrangement, in particular the pressure and other sizes in a hydraulic device, here an adjusting element for a cylinder, measured and can be influenced by specifications of the control arrangement. This influencing takes place in a manner known per se with one upstream of the hydraulic device continuously adjustable control valve, its valve spool  for example, is adjusted electromagnetically. Known are also electro-hydraulic pressure control circuits se with a continuously adjustable pump as an actuator.

The valves or pumps mentioned above can be used a so-called fieldbus connection and digital executed controllers such. B. a digital pressure regulator be equipped. This digital connection and signal influence in the control loop offers compared to the analog Valves new possibilities in terms of diagnostics ability, ease of use and a qualitative better control quality and the possibility of new flexible Concepts based on special properties in the system can be placed.

Are known from the prior art mentioned at the outset also replacing control arrangements for hydraulic Zy Linder drives in which two regulations such. B. Speed and pressure control can be combined can. The problem with these regulations is that Detection of the transition between the individual regulations species. For example, a switchover of the Pressure setpoint depending on the specified threshold value or z. B. also depending on a cylinder position tion of the actuating cylinder.

Prerequisite for the above possibilities of flexible control concepts with digital controllers that be designed for special properties in the system can, but are usually methods of identification the specific characteristics of the control loop, which enable evaluable changes in the hydraulic control loop can also be recognized.

Examples of this are from the technical paper "Comparison of Control concepts for an electro-hydraulic pressure regulated variable pump "in the magazine" Oil hydraulics  und Pneumatik ", 36 (1992) No. 9, pages 602-613 men. Here is z. B. described how from the pump position away and the system pressure a consumer characteristic in Control loop can be identified.

Advantages of the invention

A method of operating a control arrangement for a hydraulic system in which the actual values in the hydraulic system are detected by sensors and in particular the pressure of the hydraulic fluid in the hydraulic System is regulated, is with the characteristic features len of the main claim in an advantageous manner forms. Here the actual pressure values and the deflections control signal in the hydraulic system by Sen sensors and the effective system volume of the Hydraulic fluid in the closed volume of the Pressure control system calculated from the measured values and system parameters are converted from the calculated system volume for automated controller parameterization usually order derived.

The effective system volume from the measured values is particularly advantageously determined continuously or at intervals by determining the hydraulic capacity of the controlled system during the operation of the controlled system. Individual process steps, which are explained in more detail using the exemplary embodiment shown in the figures, are:

• - The actual values of the pressure and the valve or pump penhubs as deflection of the control signal are in predetermined time intervals scanned,  
• - The sampled signals are filtered and with the pressure gradient becomes a differentiator averages,
• - The pressure gradient is detected with a threshold value gate then examines whether in a given Time there is a sufficient pressure change,
• - If the pressure change is smaller than the specified threshold value, the hydraulic capacity (C _H ) is set to zero,
• - If the pressure change is greater than the specified threshold value, the volume flow is calculated, then constant with approximately the same filter time as in the differentiating element, and the hydraulic capacity (C _H ) is calculated from the values determined in this way.

With knowledge of the effective system volume determined in this way is the most important system parameter in a simple way meters for parameterizing a pressure regulator supply. A change in this parameter over a longer period Period allows conclusions to be drawn about the system properties such as B. a leak in the hydraulic system and can can therefore be used advantageously for diagnostic purposes. Be It is particularly advantageous that the identification of the sys temperature parameters carried out in a closed control loop can be, so the identified size according to very a short time, e.g. B. almost at the beginning of the control process, is available with high accuracy and robustness. It is also advantageous that the command variable of the Re gelkreises have any signal behavior here may be, for example, a rectangle, triangle or sine that the direction of the signal change (Dru purchase or reduction) has no influence on the result and that the computational effort is relatively low, so that  online identification is also possible during operation is.

According to an advantageous embodiment, the iden system volume of the hydraulic pressure control stretch with artificial neural during operation Networks (KNN) or evaluated using fuzzy logic and classified so that the system parameters of the Control arrangement are easy to determine. With the neurona len networks or fuzzy logic result e.g. B. Errors in the identification, which are less than 5%. With egg appropriate optimization or a higher acceptance dance is also possible to reduce the computing effort Lich.

In one embodiment of the method according to the invention The system parameters for parameterizing the Control arrangement in a particularly advantageous manner for Er averaging a changeover time for the changeover from speed regulation or control to egg ner pressure or force control in a control arrangement range covered. In this embodiment of the fiction According to the procedure, it can be ensured that during the pressure control the pressure control Speed control process by their intervention only then bothers when the pressure conditions in the hydraulic System actually cause critical operating conditions or if the desired guidance in the pressure control process behavior is achieved.

The process of switching can be more advantageous here Way done automatically, the process both for drives with a valve as well as with an electro hydraulically adjustable pump is suitable. On simple che way, the switching time can be determined from one of the Valve stroke of the actuator calculated volume flow and  a volume flow calculated from the pressure gradient be determined, the difference between the two volumes currents represents a measure of the changeover time.

The changeover time can advantageously be determined using the following method steps:

• - The actual values of the pressure and the valve or pump penhubs as deflection of the control signal are in predetermined time intervals scanned,
• - the pressure gradient is determined by differentiating telt, the volume flow is on the one hand from the pressure graph served calculated and
• - The volume flow is on the other hand from the valve or pump stroke calculated.

In the solution known from the prior art for example the threshold values to be set here the switching greatly from the operating conditions such as z. B. depend on the load mass or the travel speed gig, so a considerable hassle of hiring these values when the operating conditions change necessary is. The method according to the invention is there against largely independent of these operating conditions because it is based on the fact that the pressure control at a completed hydraulic volume and it uses the existing ones in electrohydraulic drives Measurement signals such as pressure and actuator stroke for calculation of the changeover time.

An advantageous application of the method according to the invention rens results in particular with a control arrangement, in which an electro-hydraulic pressure control circuit with egg a continuously adjustable valve or a continuously adjustable valve adjustable pump, e.g. B. for an actuating cylinder Injection molding machine, is equipped. These are by means of  a digital fieldbus connection with a digital one Controller connected, the calculation of the Sys can be carried out using a software module is.

These and other features of preferred training gene of the invention go out not only from the claims the description and the drawings, the individual characteristics individually or for more ren in the form of sub-combinations in the execution form of the invention and realized in other fields his and advantageous as well as protectable execution represent representations for which protection is claimed here becomes.

drawing

Exemplary embodiments of a control arrangement for explanation of the method according to the invention are based on the drawing explained. Show it:

Fig. 1 is a schematic diagram of a control arrangement for a hydraulic system having a continuously changed derbaren valve,

Fig. 2 is a flowchart for calculating the hydrauli rule capacity of the hydraulic system in the Regelan order of FIG. 1,

Fig. 3 is a diagram of the course of the pressure in Hyd rauliksystem and the path of an actuator of the control device over time,

Fig. 4 is a diagram of the course of ten Druckgradien and the volume flow in the hydraulic system over time;

Fig. 5 is a diagram of the course of calculated from the values of FIG. 1 hydraulic capacity over time,

Fig. 6 is a diagram of the course of the calculated hydraulic capacity over time with difference comprehensive system volumes,

FIGS. 7 and 8 is a diagram of the course of calcu Neten hydraulic capacity over time in dependence from the pressure level and the direction of egg ner pressure change,

Fig. 9 is a graph of the change of Ersatzkompres sion modulus of the hydraulic fluid depending on the pressure,

Fig. 10 is an illustration of the normalized input values of the hydraulic capacity as features for identification of the system volume in the hydraulic system,

Fig. 11 is a diagram of the course of pressure over time at a start of an actuating cylinder against a stop,

Fig. 12 is a diagram of the course of the stroke of the actuating cylinder over time at a start of the control cylinder against a stop,

Fig. 13 is a diagram showing a comparison of flow rates in the hydraulic system on the one hand determined from the measured pressure and on the other hand from the measured stroke of the actuating cylinder who to,

FIG. 14 is a diagram of the course of pressure over time at a start of an actuating cylinder at a stop for determining a switching time point in the control method;

Figure 15 is the difference between the volume flows procedure. A diagram of the curve of FIG. 13, taking into supply a threshold value as a measure of a switching and

Fig. 16 is a flow chart of the calculation steps for the detection of the switching timing of FIGS. 14 and 15.

Description of the embodiments

In one embodiment, a control arrangement of FIG. 1 it is shown that the inventive method can for example be used to control a hydraulic Stellzy Linders for an injection cylinder in a plastic injection molding machine 1. In order to ensure the high quality of the injection molded part produced, the injection speed of the liquid plastic, the holding pressure and the dynamic pressure during the molding process must be controlled as well as possible, even if the molding compound partially cools.

In the control arrangement according to FIG. 1, the actual pressure value p in the volume flow of the hydraulic fluid is recorded with a suitable sensor arrangement 2 and converted into a preferably digital electrical signal by means of a signal converter 3 . In a controller 4 , which is only indicated schematically here, the measured values and the target values for the pressure and the volume flow of the hydraulic fluid are evaluated and then fed to a continuously adjustable valve 5 . The other components, which are not explained in greater detail here, correspond to the known structure of a hydraulic control circuit.

With the arrangement shown here, the effective system volume of the hydraulic fluid in the closed volume of the pressure control system can thus be calculated from the measured values and the system parameters for parameterizing the control arrangement can be derived from the calculated system volume. Two measurement signals are available to identify the system volume, namely the controlled variable pressure and the deflection of the actuating cylinder. The effective system volume is determined from the measured values by determining the hydraulic capacity C _{H of} the controlled system during operation.

In Fig. 2, the calculation process for determining the hydraulic capacity C _{H is shown} contiguously using a flow chart. With the help of the measured pressure signal p and the likewise detected displacement signal s, it is possible to use the hydraulic capacity C _H , which is defined as the quotient of the volume flow Q and the pressure gradient dp / dt () for a pressure control system with closed volume, according to the following formula to calculate:

with E corresponding to the elasticity or better Er set compression module and Vo according to the system volume hydraulic fluid. With the replacement compression mo dul E for the hydraulic fluid becomes a standard here assumed constant value of a certain usual air fraction and other known influencing factors includes.

The volume flow Q or Q ₁ results for a valve with a linear characteristic in a constant pressure system depending on the stroke s of the actuator, derived from the flow law through an orifice according to Bernoul li, by the relationship:

for s ≧ 0 there is Δp = p ₀ - p _A and
for s ≦ 0 there is Δp = p _A - p _T.
wherein Ap _nominal, the nominal pressure difference, Q _nominal current the rated volume, p is ₀ the system pressure p _{A is} the pressure at the Stellzylin, pr the tank pressure in the hydraulic system and the s. Ven tilhub is based on the nominal stroke, here the nominal volume flow Q _{nenn is known} as a value already determined by the manufacturer of the hydraulic system.

If, instead of the valve, there is a pump as an actuator, the volume flow Q _{1 can} be determined in the following way:

Q ₁ = nV _{funding S} - Q _leak (3),

wherein the pump speed, V n the _conveyor per revolution, to the nominal stroke, and Q represents _leakage s the pump stroke relative the leakage of the hydraulic fluid. The leakage Q _{leak of} the pump is approximately linearly dependent on the pressure p in the following way:

Q _leak = factor.p (4)

Another possibility for calculating the volume flow Q or Q ₂ is obtained by using the pressure signal s p. The hydraulic capacity C _H is defined here as:

E _G as the equivalent bulk modulus, Q _to as the flowing to flow and Q _from the effluent as Volu volume flow in the hydraulic system.

In a closed volume of the effluent flow rate Q can be set _from zero, so that after a changeover of the formulas from the previous equation gives the following:

The pressure gradient can be determined by differences adorn the pressure signal p, as described below, determine.

According to the flow chart from FIG. 2 with values given here only for example, to suppress interference, the measurement signals p and s are first sampled and filtered in a second-order low-pass filter. After differentiating the pressure signal p to determine the pressure gradient, a threshold value detector detects the time in which a sufficient pressure change (eg <200 bar / s) takes place in the system. If the pressure change is less than the predetermined threshold value, the hydraulic capacity C _H is set to zero as a result.

However, if the threshold value is exceeded, the volume flow Q is calculated according to equation (2) or (3). In order to exclude a phase shift between the volume flow Q or Q ₁ or Q ₂ and the pressure gradient, the volume flow Q is subsequently filtered again with the same filter time constant (PT1) as used in the previously mentioned differentiating element (DT1). Then the hydraulic capacity C _{H can} be calculated.

With reference to FIG. 3 waveforms are exemplary p for printing and lift s shown at a pressure jump from 40 to 140 bar. In FIG. 4, from the dimensional Messsig calculated values for the volume flow Q and the pressure gradient is removed. From Fig. 5, the course of the deduced hydraulic capacity C _H, as in the previous cases with respect to time t, recognizable.

Fig. Shows how the hydraulic capacity C t _H behaves as a function of different system volumes over time. 6 One can see here a strong dependency of the C _H level on the volume Vo of the hydraulic system.

From Fig. 7 it can be seen that the curve shape of the hydraulic capacity C _H not only depends on the volume Vo of the hydraulic system but also on the level of the pressure change or on the direction of the signal change (pressure / reduction), which in particular also from the 8 is in the illustrations of FIG. seen in the left pressure increase from 190 to 280 bar and in th computational graph, a pressure reduction from 190 to 100 bar is ge shows. This influence can largely be attributed to the change in the replacement compression module E _G as a function of the pressure p.

In Fig. 9, the replacement compression module E _G in dependence on the pressure p and the percentage (0.1 to 10%) of undissolved air in the hydraulic fluid is Darge.

In order to take into account the influences of the pressure level and the signal direction when identifying the system volume, the time period Δt can be chosen to be very short after the pressure gradient threshold value has been exceeded for calculating the hydraulic capacity C _H. As a guideline for pressure control with a pump z. B. a time period of 8 ms has been found to be sufficient. This results in a total duration for the entire pressure jump of approximately 100 ms, ie after 8% of the total duration of a leadership jump, sufficient information is already available to identify the system volume.

With a sampling rate of the digital pressure regulator according to the exemplary embodiment of approx. 200 μs, 41 C _H values are available after approx. 8 ms, whereby it has proven to be sensible to select 5 of the 41 values that are constant while the others are discarded can be. In addition to the 5 C _H values, the pressure p at the start of the calculation and the pressure p after 8 ms are saved. With the information of these two pressure values, both the jump direction and the pressure level can then be recognized. After normalizing the 5 C _H values and the pressure values p, there is a characteristic for identification, as shown in FIG. 10 for different volumes.

The characteristic developed in this way for identifying the system volume of a hydraulic pressure control system can be evaluated using various classification methods, including so-called artificial neural networks (KNN) or using so-called fuzzy logic. With the artificial neural networks arise z. B. Error in the identifi cation less than 5%. The number of feature values of 7 as normalized input values indicated in the diagram according to FIG. 10 is to be seen as a guide value. With a corresponding optimization or acceptance of larger errors, fewer values and thus less computing effort are possible.

A change in the replacement compression module E _G depending on the temperature, the hose material, the proportion of undissolved air, etc. in the hydraulic system has an effect on the level of the C _H curves shown in FIGS . 3 to 8 and thus on the identified system volume , However, this is desirable since the behavior of the control path also changes as a result and can be well described with the identified volume value changed in the same ratio. The identified system volume can also be understood as a reference variable for the system for which the characteristics were created.

An increased leakage in the system affects e.g. B. at iden system volume so that different Values for pressure build-up or pressure reduction calculated become. This can then be used to diagnose the leak in the hyd system can be used.

The course of the replacement compression module E _G according to FIG. 9 can be described very well with an air fraction of 0.2% for most applications. The system volume required for the calculation can then either be determined from the identification of the system parameters explained above or be adapted so that the volume flows Q ₁ (equation (2)) and Q ₂ (equation) calculated from the actuator stroke s and from the pressure gradient chung (6)) show the same behavior when the volume is closed.

Fig. 11 and Fig. 12 show the signal behavior of the pressure (Fig. 11) and valve lift (Fig. 12) at a toggling control of a speed control in a force control on a cylinder drive of an injection molding machine according to the Fig. 1, as based in principle the prior art has already been mentioned in the introduction to the description. The cylinder of the actuator runs here according to the diagram of FIG. 12 with a valve opening of approximately 25% against a stop. When the stop is reached there is a closed volume in the hydraulic system. Due to the large valve opening, the pressure p rises very quickly, as can be seen in FIG. 11, on the large piston surface of the Stellzylin and must be checked by switching on the force regulator in good time, ie the valve lift is drastically reduced.

The time of the previously described transition from speed control to force control can be detected with the above-described method according to the invention, in some cases with a corresponding switchover concept. To illustrate, in Fig. 13 of the valve from the calculated flow rate Q s ₁ calculated from the pressure gradient flow rate Q ₂ represents gegenü mountains. After the actuating cylinder has moved to the stop and a closed volume is present, the volume flow Q ₂ rises sharply and approaches the volume flow Q ₁ very quickly. It can be clearly seen here that the two volume flows Q ₁ and Q ₂ calculated using equations (2) and (6) show a very good agreement when the volume is closed.

The difference between the two volume flows Q ₁ and Q ₂ according to FIG. 13 is then a measure of the actual switching process. Is shown in FIG. 14 and 15, the difference is greater than the threshold value SW, here. B. -0.6.10 ^-4 m ³ / s, the changeover time UZ is detected precisely in the rise phase of the pressure p, so that the force regulator can be switched on in good time.

The calculation steps are shown over time for detection of the switching timing UZ from the flowchart of Fig. 16. After scanning the he summarized pressure signal p and the stroke signal s becomes He averaging the flow rate Q _2, the pressure signal at their differen p and the volume flow Q ₂ is then calculated from the pressure gradient. In parallel, the volume flow Q _{1 is} calculated from the stroke signal s. The evaluation of the difference of these volume values Q ₁ and Q ₂ takes place via the threshold value SW, as described with reference to FIGS. 14 and 15.

Claims (9)

1. Method for operating a control arrangement for a hydraulic system, in which
the actual values in the hydraulic system are recorded by sensors ( 2 ) and in particular the pressure (p) of the hydraulic fluid in the hydraulic system is regulated, characterized in that
the effective system volume of the hydraulic fluid in the closed volume of the pressure controlled system is calculated from the measured values and that
system parameters for parameterizing the control arrangement can be derived from the calculated system volume. 2. The method according to claim 1, characterized in that the effective system volume from the measured values th by determining the hydraulic capacity (C _H ) of the controlled system during operation of the controlled system is determined continuously or at intervals (Δt). 3. The method according to claim 2, characterized in that
the hydraulic capacity (C _H ) is determined using the following procedure steps:
the actual values of the pressure (p) and the valve or pump stroke (s) as the deflection of the actuating signal who are sampled at the predetermined time intervals (Δt),
the sampled signals are filtered and the pressure gradient (p) is determined using a differentiator,
the pressure gradient (p) is examined with a threshold value detector to determine whether there is a sufficient change in pressure within a predetermined time,
if the pressure change is less than the specified threshold value, the hydraulic capacity (C _H ) is set to zero,
If the pressure change is greater than the specified threshold value, the volume flow (Q; Q ₁ , Q ₂ ) is calculated, then filtered with the filter time constant (PT1) which is approximately the same as in the differentiating element (DT1) and the values determined in this way become the hydraulic one Capacity (CH) calculated. 4. The method according to any one of the preceding claims characterized by that the identified system volume of the hydraulic Pressure control system during operation with artificial neural networks or using fuzzy logic is evaluated and classified so that the system pa parameters of the control arrangement can be determined.   5. The method according to claim 1, characterized in that the system parameters for parameterizing the control order for determining a changeover time (UZ) for switching from one speed speed regulation or control to a pressure or Force control used in the control arrangement become. 6. The method according to claim 5, characterized in that the switching time (UZ) on the basis of a Ven tilhub (s) of the actuator ( 1 ) calculated volume flow (Q ₁ ) and a calculated from the pressure gradient th volume flow (Q ₂ ) is, the difference between the two volume flows (Q ₁ , Q ₂ ) represent a measure of the changeover time (UZ). 7. The method according to claim 6, characterized in that
the changeover time (UZ) is determined using the following procedural steps:
the actual values of the pressure (p) and the valve or pump stroke (s) as the deflection of the actuating signal who are sampled at predetermined time intervals (Δt), the pressure gradient (p) is determined by differentiating,
the volume flow (Q ₂ ) is calculated from the pressure gradient (p) and
the volume flow (Q ₁ ) is calculated from the valve or pump stroke (s). 8. Control arrangement for performing the method according to one of the preceding claims, characterized in that an electrohydraulic pressure control circuit is equipped with a continuously adjustable valve ( 5 ) or a continuously adjustable pump, which are connected to a digital controller by means of a digital fieldbus coupling, wherein the calculation of the system volume can be carried out in the controller by means of a software module. 9. Control arrangement according to claim 5, characterized in that with the continuously adjustable valve ( 5 ) or the ste adjustable pump, the hydraulic actuating cylinder of an injection molding machine ( 1 ) is controllable.